Learning-based estimation of dielectric properties and tissue density in head models for personalized radio-frequency dosimetry
This work addresses the bottleneck of inter-variability assessment in radiation safety for personalized dosimetry by enabling faster model generation without segmentation.
The study tackled the time-consuming process of creating personalized head models for radio-frequency dosimetry by developing a learning-based approach that directly estimates dielectric properties and tissue density from MRI images, achieving highly consistent specific absorption rate (SAR) distributions compared to conventional segmentation-based methods.
Radio-frequency dosimetry is an important process in human safety and for compliance of related products. Recently, computational human models generated from medical images have often been used for such assessment, especially to consider the inter-variability of subjects. However, the common procedure to develop personalized models is time consuming because it involves excessive segmentation of several components that represent different biological tissues, which limits the inter-variability assessment of radiation safety based on personalized dosimetry. Deep learning methods have been shown to be a powerful approach for pattern recognition and signal analysis. Convolutional neural networks with deep architecture are proven robust for feature extraction and image mapping in several biomedical applications. In this study, we develop a learning-based approach for fast and accurate estimation of the dielectric properties and density of tissues directly from magnetic resonance images in a single shot. The smooth distribution of the dielectric properties in head models, which is realized using a process without tissue segmentation, improves the smoothness of the specific absorption rate (SAR) distribution compared with that in the commonly used procedure. The estimated SAR distributions, as well as that averaged over 10-g of tissue in a cubic shape, are found to be highly consistent with those computed using the conventional methods that employ segmentation.